Breakout + Spread!! Sure!!! | Page 14 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Should Trade with Broker offering low spread or wide spread? 12 replies

Variable spread vs FIXED spread & broker comment/feedback 6 replies

ECN, fixed spread or variable spread account? why? 1 reply

Oanda spread versus MB Trading spread 27 replies

1pips spread is showing as 18 spread, anyway to overcome? 1 reply

Attachments: Breakout + Spread!! Sure!!!
Exit Attachments

Breakout + Spread!! Sure!!!

  • #261
  • Jul 20, 2007 12:58pm Jul 20, 2007 12:58pm
Quoting hany mohssen
Hi all

So the TP s 10Pips+spread and stoploss ???


which is the best paires for this System?

Regards
Eagle originally said the SL is 15 pips.
He also said the best pairs are GBP/USD , GPP/JPY

This was in one of his early posts.
  • #262
  • Jul 20, 2007 1:00pm Jul 20, 2007 1:00pm
Quoting billbss
Eagle originally said the SL is 15 pips.
He also said the best pairs are GBP/USD , GPP/JPY

This was in one of his early posts.
but i think that SL 15 will be bad?

Regards
  • #263
  • Jul 20, 2007 1:03pm Jul 20, 2007 1:03pm
Quoting hany mohssen
but i think that SL 15 will be bad?

Regards
I'm not sure.

We are doing some backtesting. There should be some results soon.

Eagle also just posted an indicator. He said it will help with the SL. I haven't looked at it yet, so I don't know how it works.
  • #264
  • Jul 20, 2007 1:26pm Jul 20, 2007 1:26pm
Quoting billbss
I'm not sure.

We are doing some backtesting. There should be some results soon.

Eagle also just posted an indicator. He said it will help with the SL. I haven't looked at it yet, so I don't know how it works.
could you help me to get this Indicator co i can't found it
the last post for Eagle is the Bout indictor with a diff. time

Regards
  • #265
  • Jul 20, 2007 1:41pm Jul 20, 2007 1:41pm
Quoting hany mohssen
could you help me to get this Indicator co i can't found it
the last post for Eagle is the Bout indictor with a diff. time

Regards
You are right. That is not the indicator that I assumed it was.

Eagle stated the SL was 15 pips. We are doing some backtesting to see if that can be improved.
  • #266
  • Jul 20, 2007 3:01pm Jul 20, 2007 3:01pm
Guys

Have been watching this thread with some interest.

I think you can filter a lot of false signals using a stochastics oscillator set to 5,3,3,close and only enter long when signal is over 80 / short when under 20.

It might mean you delay entry until will into the day but a quick scan back on GBPUSD and this looks like it has potential.

All the best

Jason
  • #267
  • Jul 20, 2007 3:11pm Jul 20, 2007 3:11pm
Quoting aeu97137
Guys

Have been watching this thread with some interest.

I think you can filter a lot of false signals using a stochastics oscillator set to 5,3,3,close and only enter long when signal is over 80 / short when under 20.

It might mean you delay entry until will into the day but a quick scan back on GBPUSD and this looks like it has potential.

All the best

Jason
Thanks for contributing to the group effort . Hopefully, on Monday we will have some results of back testing to see which time frame works best.
  • #268
  • Jul 20, 2007 4:37pm Jul 20, 2007 4:37pm
Attached Image
  • #269
  • Jul 21, 2007 11:46pm Jul 21, 2007 11:46pm
Hi guys, for a buy situation, do you wait for the bid or the ask price to pass the breakout zone?

thanks.
  • #270
  • Jul 22, 2007 4:06am Jul 22, 2007 4:06am
Somebody try to backtest? for me this dont works..

http://i20.servimg.com/u/f20/11/12/22/22/chart10.gif


http://i20.servimg.com/u/f20/11/12/22/22/tester10.gif
  • #271
  • Jul 22, 2007 8:47am Jul 22, 2007 8:47am
Ledzep: if you want to test a system with such narrow stops and targets like 10 to 15 pips, the software has to account for the bid/ask-spread, which, to my knowing, metatrader cannot do in a feaseable way.

I have some similar approaches already coded in my own custom written backtester, so i ran this system through 3 years of oandas tickdata, since it wasnt much work for me. The result is catastrophic. Settings i used: 10 pips target, 15 pips stop, range from 6 to 9 CET. Trading was limited to 9-13 CET. Times adjusted on CET-DST rules. If anyones interested, i can post the system-stats and transaction-list.

On a side note, all cable morning-breakout approaches i track seem to perform really bad since Jan 2007.
  • #272
  • Jul 22, 2007 8:58am Jul 22, 2007 8:58am
shrike:

What software do yo use to backtest oanda tick data?
  • #273
  • Jul 22, 2007 9:04am Jul 22, 2007 9:04am
Quoting shrike
Ledzep: if you want to test a system with such narrow stops and targets like 10 to 15 pips, the software has to account for the bid/ask-spread, which, to my knowing, metatrader cannot do in a feaseable way.

I have some similar approaches already coded in my own custom written backtester, so i ran this system through 3 years of oandas tickdata, since it wasnt much work for me. The result is catastrophic. Settings i used: 10 pips target, 15 pips stop, range from 6 to 9 CET. Trading was limited to 9-13 CET. Times adjusted on CET-DST rules. If anyones interested, i can post the system-stats and transaction-list.

On a side note, all cable morning-breakout approaches i track seem to perform really bad since Jan 2007.
I'm trying to backtest it on AmiBroker and am getting very bad results with it, also.

I'll double check my code before I post the results to make sure I haven't made a mistake.
  • #274
  • Jul 22, 2007 9:10am Jul 22, 2007 9:10am
I wrote my own software in VB.NET. My program chews either thru tick-files or realtime tick-streams and outputs transaction-lists. I use neoticker to analyze those transaction-lists and generate stats/equity-curves, and another program that i wrote myself to analyze P/L-distributions.
  • #275
  • Jul 22, 2007 9:24am Jul 22, 2007 9:24am
Quoting shrike
I wrote my own software in VB.NET. My program chews either thru tick-files or realtime tick-streams and outputs transaction-lists. I use neoticker to analyze those transaction-lists and generate stats/equity-curves, and another program that i wrote myself to analyze P/L-distributions.
Your testing is much more sophisticated than mine.

I'm using Alpari 1M bars with AmiBroker.

BTW- I do know of a GBP breakout system that has done really well this year. It's hitting over 80% winners. Profit/ risk is 20/30, but it tests even better with a 35/30.

Someone was running a journal thread on this system a couple of months ago and the Mod deleted it. I don't want to say anymore about it because of that.
  • #276
  • Edited 10:35am Jul 22, 2007 10:19am | Edited 10:35am
so we have 3 different software packages

1 MT4 using Breakout eagle Ea
2 Visual basic using Oanda
3 AmiBroker using Alpari

All three are returning poor results. Is that correct?
  • #277
  • Jul 22, 2007 10:33am Jul 22, 2007 10:33am
smjones: i am not entirly clear what stoploss eagle uses and what timerange, so his system may be quite different from what i did tested.
  • #278
  • Jul 22, 2007 11:17am Jul 22, 2007 11:17am
Quoting shrike
smjones: i am not entirly clear what stoploss eagle uses and what timerange, so his system may be quite different from what i did tested.
I think this is one thing that no one is clear about.
Looking at some of the charts he has posted recently, some are using 5-8am GMT and others 4-7am. So I don't know if he uses a different time range for different currency pairs or what.
Then I also saw the 6-9am range indicator he has posted a couple of times and that really throws me off. (on a side note, through the little testing I have done on my own the 6-9am range does not do well)

The other thing I wanted to point out is that I think the targets (both profit and stop loss which from what I understood are 10 and 15 pips) are not the most efficient we can use. Sure you might get a higher win %, but that is irrelevant. Because a 15 pip stop loss plus spread (say 18 pips total for the GBPUSD) is almost double the 10 pip profit target. And more often than not, these breakouts go for much more than 10 pips. I think by increasing the profit target to say 20, it would really increase the expectancy of the system overall. Win % will go down, but profit will increase.

Anyway, I'm still doing my own testing with different pairs and time ranges. I will keep everyone posted of my findings. But so far, I wouldn't call this system dead yet based on backtesting.

Tom
  • #279
  • Jul 22, 2007 11:35am Jul 22, 2007 11:35am
I have tried all possible time ranges with poor results. I was using SL 15 and PT 10.

Optimizing the SL and PT using 1M bars is very slow. I have not completed that yet.

From what I've seen so far, I'm almost certain it will be a loser no matter what parameters are used.
  • #280
  • Jul 22, 2007 11:37am Jul 22, 2007 11:37am
Quoting shrike
smjones: i am not entirly clear what stoploss eagle uses and what timerange, so his system may be quite different from what i did tested.
Eagle uses 10 pips + spread for PT. His SL is up in the air. I think lets it run and uses a discretionary exit.

We've nailed his time range to 6-9 AM London time.
0 traders viewing now